Abstract: Determination of nitrogen isotopes in natural samples plays an important role in scientific exploration of the global nitrogen cycle. However, compared with carbon isotope measurement, the accuracy of nitrogen isotope ratio determination is generally lower when using the high-temperature combustion method (CM) based on an elemental analyzer coupled with an isotope ratio mass spectrometer (EA-IRMS). This study optimized the reaction conditions and parameters in terms of combustion conditions, different reaction tube configurations, and interfering gases in the elution gas flow. The accuracy of nitrogen isotope testing in the EA Isolink-IRMS system was compared under different reaction tube configuration (single and double modes) and oxygen injection parameter conditions. By measuring a series of standard substances and marine sediments with varying nitrogen contents, the method for nitrogen isotope measurement was further optimized. The results showed that different reaction tube configurations, sample combustion states, and filler consumption in EA-IRMS have significant impacts on the determination of nitrogen isotope ratios. Accurate nitrogen isotope analysis requires ensuring sufficient sample combustion in EA, sufficient reduction of nitrogen oxides, and removal of CO₂ and potential impurity gases from the gas stream. For the determination of nitrogen isotopes using the EA Isolink-IRMS system, this study showed that: 1) Depending on the sample type, the appropriate oxygen injection amount should be selected by rapid oxygen injection to ensure sufficient sample combustion. For small-sized samples with high nitrogen content (e.g., plants, animal tissues, and pure chemical substances), an oxygen flow rate of 175 mL/min and injection duration time of 3 s should be used. For large-sized samples (e.g., marine sediments with low nitrogen content), the oxygen flow rate of 250 mL/min and the injection duration time of 5 s were recommended. 2) Based on the current single tube configuration, adding a reduction tube with smaller volume can effectively ensure the full reduction of nitrogen oxides and absorption of excess oxygen. When using the optimized method for nitrogen isotope measurement, a signal intensity of approximately 50 mV can be generated for every 1 μg of N in the sample. The standard deviations of the measured δ¹⁵N values of reference materials (USGS40, USGS65 and USGS64) across different batches are better than ±0.15‰. For marine sediment samples, the measurement accuracy of nitrogen isotopes can be better than 0.10‰ in a batch. The stability, reproducibility, and accuracy of the measurements can be greatly improved. This method provides crucial technical support for the determination of nitrogen isotope ratios.